Raven.io (@Ravencloudinc) header
Raven.io (@Ravencloudinc) profile picture
Palo Alto, CA
Joined May 2023
Photo by Raven.io (@Ravencloudinc)Photo by Raven.io (@Ravencloudinc)Photo by Raven.io (@Ravencloudinc)Photo by Raven.io (@Ravencloudinc)Photo by Raven.io (@Ravencloudinc)Photo by Raven.io (@Ravencloudinc)Photo by Raven.io (@Ravencloudinc)Photo by Raven.io (@Ravencloudinc)Photo by Raven.io (@Ravencloudinc)Photo by Raven.io (@Ravencloudinc)
👀 When your application hasn’t started, but the attack has.
1
16
Recent ARTEX campaigns show AI agents exploiting application flaws, reusing public exploits, and moving from initial access to root in minutes. Runtime application context critical. Security teams need to know which code is actually executing, what it is doing, and what it should be allowed to do. Read the full breakdown: na2.hubs.ly/H08_CQ20 #ApplicationSecurity #RuntimeSecurity #AIsecurity
1
16
🍳 AI is speeding up the time it takes an attacker to ‘cook an exploit’
1
2
12
🩹 Patching faster is not a security strategy.
1
22
How ready are you for an AI driven attack? If a risky package reaches production, can your team spot it and act in time? See how prepared you are to identify what’s running, assess your exposure, and respond to an application attack with Raven’s ADR Readiness Assessment. 11 questions. About 3 minutes. A clearer view of your readiness. na2.hubs.ly/H08N8Nv0
15
A patching strategy alone does not cut it anymore. James Berthoty of Latio breaks down why modern application risk moves faster than traditional patch cycles, especially as open-source software and AI accelerate vulnerability discovery and exploitation. Security teams still need to patch, but they also need a way to understand and reduce risk while the application is running. Watch the full webinar on demand: na2.hubs.ly/H08N8F30 #ApplicationSecurity #RuntimeSecurity #VulnerabilityManagement #AppSec
1
24
A known PeopleSoft flaw, Java deserialization, and an unpatched system were enough to expose sensitive FBI data. The bigger lesson is not just “patch faster.” It is that once malicious code reaches the application runtime, security teams need visibility into what actually executed and the ability to stop it there. Read the breakdown: na2.hubs.ly/H08zY_C0
11
Node2Shell is here. Right after PHP2Shell, we now have a high-severity RCE in http://Node.js itself: CVE-2026-111497. A crafted Brotli stream can trigger an out-of-bounds write in the decoder that ships with http://Node.js. It is reached automatically — through Express/body-parser on incoming Content-Encoding: br requests, or through the built-in fetch() on responses. No application bug required. The runtime is the attack surface. This is exactly the kind of capability Raven mapped in The 14 Most Powerful Families in Software. The Brotli decoder sits at the intersection of the Customs Officer (parsing untrusted input) and the Smuggler (native code), fed by the Doorman (Express). These libraries disappear at runtime. Attackers already know the path. Patch to 22.23.4 / 24.22.1 / 26.12.1. Then ask the harder question: what else is executing that you cannot see? Read the technical analysis: na2.hubs.ly/H08zYS60 Download the Raven research: na2.hubs.ly/H08zYS70 #NodeJS #ApplicationSecurity #RuntimeSecurity #OpenSourceSecurity
114
Not every critical vulnerability is equally critical in your environment. Runtime evidence shows which libraries are actually loaded, which execute, and whether the vulnerable function itself ran. That gives AppSec teams a much clearer way to decide what deserves attention first. See how runtime context can sharpen vulnerability prioritization: na2.hubs.ly/H08xhqX0 #ApplicationSecurity #RuntimeSecurity #VulnerabilityManagement #AppSec
9
SOC teams need application level context.  Watch as Omer explains how runtime application context connects suspicious behavior back to the code itself, including the library, function, and execution path responsible. That context can help SOC teams investigate faster, reduce guesswork, and understand whether an event represents real application risk. Watch the full webinar on demand: na2.hubs.ly/H08xh730 #RuntimeSecurity #SOC #ApplicationSecurity #ADR
1
1
2
134
The runtime is part of your attack surface. PHP2Shell just proved it. Two new PHP vulnerabilities (CVE-2026-109618 and CVE-2026-109619) can allow unauthenticated remote code execution. On Apache mod_php, exploitation can start before your application code gets control. No unsafe upload handler required. No session_start() required. Just PHP parsing a crafted request and mishandling an out-of-memory failure. With PHP behind roughly 70% of sites with a known server-side language, this one is worth your attention today. ✅ Update to 8.5.12, 8.4.27, 8.3.36, or 8.2.35 ✅ Inventory the runtime and its config, not just your packages ✅ Look for execution evidence, not just request logs Full breakdown from our team: na2.hubs.ly/H08xRDV0 #PHP #AppSec #RuntimeSecurity #CyberSecurity
111
Open-source attacks move faster than most AppSec programs can respond. When a vulnerable or malicious package reaches production, the question becomes whether your team can see what is actually running, understand the real exposure, and respond before it turns into an incident. The ADR Readiness Assessment helps benchmark how prepared your security program is for that reality. 11 questions. About 3 minutes. Take the assessment: na2.hubs.ly/H08lK170
1
19
AI is killing SCA and ASPM as we know them. AI-generated code is creating security risks that may never get a CVE. No advisory. No severity score. No finding to prioritize. For CISOs, that raises a question: What does your security strategy look like when there’s no CVE to act on? SCA and ASPM help us manage known vulnerabilities. But prioritization can’t protect us from risks that never enter the queue. We need to ask more than “What do we need to patch?” What should the code running in our applications be allowed to do? Should a third-party library be able to launch a shell? Read sensitive files? Make an unexpected outbound connection? We can set and enforce those boundaries before a CVE exists. Patching and prioritization remain essential. But protection cannot wait for disclosure. In my latest blog, “AI Is Killing SCA and ASPM as We Know Them,” I explore what comes next for third-party code security. na2.hubs.ly/H08kMS_0
36
The vulnerability apocalypse is coming. Anthropic just announced free AI-powered vulnerability scanning for open-source maintainers. Now imagine thousands of projects opting in. An explosion of vulnerabilities. Overwhelmed maintainers. Security teams drowning in CVEs. Enterprises unable to patch fast enough. AI is accelerating vulnerability discovery faster than organizations can remediate. And attackers don't need to wait for a CVE anymore. Why should your security strategy depend on one? Raven’s latest perspective on the future of application security. Read the full blog: na2.hubs.ly/H08kC3P0
1
45
🚨 Last chance to register. We’re live tomorrow. Join Raven and James Berthoty of Latio for Before the CVE: The 14 Software Families AI Will Target First. We’ll reveal original research, explore how AI is accelerating pre-CVE attacks, and demonstrate live exploits traced to the exact open-source library, function, input, and runtime call chain responsible. 📅 Tomorrow, October 8 ⏰ 12 PM ET / 9 AM PT 🔗 na2.hubs.ly/H08fd710 Can’t attend live? Register anyway and we’ll send you the recording.
1
46